소개글
멀티미디어 신호처리 특론 수업때 했던 음성인식 관련 프로젝트 자료
목차
Ⅰ. Objective
Ⅱ. Speaker Recognition
Ⅲ. Overall Description of Implementation
Ⅳ. Experimental Result
Ⅴ. Conclusion
본문내용
PROTECT YOUR VALUT !
Ⅰ. Objective
You have a vault which contains your precious treasures. The vault can only be unlocked with your voice. Its security grade is not sufficient to protect your treasures from impostors. Fortunately, there is a possibility to upgrade the security software of your vault. Your objective is upgrading the software to protect your vault from impostors who have an eye on your treasures by upgrading the software.
Ⅱ. Speaker Recognition
A. Speaker Recognition
Speaker recognition is the method to find the speaker from a human voice. There are many terminologies associated with speaker recognition, but it can often divide into speaker identification and speaker verification : speaker identification is that when a voice sample is given, the system finds the speaker whose voice is identical to the given voice among the several candidates. And speaker verification is that when a voice sample is given, the system checks whether the unique speaker`s voice or not. T
참고 자료
[1] Human voice, http://en.wikipedia.org/wiki/Human_voice
[2] Cepstrum, http://en.wikipedia.org/wiki/Cepstrum
[3] Mixture Model, http://en.wikipedia.org/wiki/Mixture_model
[4] Doddington, G. R., Przybocki,M. A., Martin, A. F., Reynolds, D. A., ”"The NIST speaker recognition evaluation Overview, methodology, systems, results, perspective,”" Speech Communication, 2000, 31, 225-254
[5] Campbell, J.P., ”"Speaker recognition: a tutorial,”" Proceedings of the IEEE, 1997, 85, 1437-1462
[6] Speaker Recognition, http://en.wikipedia.org/wiki/Speaker_recognition
[7] Mel-frequency Cepstrum, http://en.wikipedia.org/wiki/Mel-frequency_cepstrum
[8] Linear Predictive Coding, http://en.wikipedia.org/wiki/Linear_predictive_coding
[9] Classification, http://en.wikipedia.org/wiki/Classification (machine_learning)
[10] Binary Classification, http://en.wikipedia.org/wiki/Binary_classification